移动机器人自主抓取作业  

Research on autonomous grasping of mobile robot

作　　者：战茜 屠大维[1,2] 

机构地区：[1]上海大学机电工程与自动化学院,上海200072 [2]上海大学上海市智能制造及机器人重点实验室,上海200072

出　　处：《计算机应用》2016年第A01期95-98,102,共5页journal of Computer Applications

基　　金：国家自然科学基金资助项目(51075252)

摘　　要：与通常生产线上固定工位的多关节机械手臂手眼协调作业机制不同,针对多自由度移动平台机载多关节机械手臂构成的移动作业机器人的自主作业问题,提出一种移动机器人自主作业研究方法。首先,机器人通过视觉传感器获取作业环境和作业对象的空间坐标及位姿信息,然后,对机器人进行整体运动学建模、运用提出的算法进行运动抓取规划,最后,将求得的结果转变为机器人移动轮子及手臂关节的输出,二者统一结合,协调完成作业任务。实验系统以机器人在局部环境中自主移动到目标并抓取作业为例,实现以目标为导向的机器人自适应抓取。实验结果表明,该自主抓取作业算法能够保证在目标位姿不同的情况下抓取作业的顺利进行,并将作业精度控制在5 mm以内,验证了该自主作业方法在移动机器人自主抓取作业中具有有效性和可行性。Different from the architecture of multi-joint robotic hand-eye coordination operation on production line,an autonomous task method was proposed in order to solve the problem for multi-degree of freedom robot with mobile platform and arm on autonomous grasping.First,positions and orientations of the operation environment and object were located by visual sensors on the robot.Then,the robotic system model was described through the kinematics of robots.The grasping plan was calculated by the proposed method.Afterwards,the result configuration was carried out through the coordinate combination of wheels and arm of the robot to complete the tasks.The experiment used a mobile robot to move to and to grasp the target automatically as an example to realize the goal-oriented and adaptive grasping based on this control method.The results show that this method can both ensure the grasping tasks whose targets have various positions and orientations go smoothly and control the accuracy within 5 mm.It is verified that the proposed method is effective and feasible in autonomous grasping of mobile robot.

关 键 词：移动机器人 机械手臂 机器人运动学 运动规划 自主抓取 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]